We submit this proposal to contribute to the development of methods for defining, measuring, and reporting on the quality of cancer care. This will support the long-term goal of providing stronger, more representative empirical evidence of process-outcome linkages in the delivery of care used to diagnose, treat, monitor, and palliate cancer patients. In this context, we propose to collaborate with other CanCORS Primary Data Collection Research (PDCR) sites and the central Statistical Coordinating Center (SCC) in the study of patients with lung and colorectal cancer. Using patient self-report data soon after diagnosis and approximately one year later, as well as medical record and cancer registry data, we will describe the relationship among patient characteristics at the time of diagnosis and relevant processes and outcomes of care; and systematically evaluate the relative contributions of the patients' demographics, comorbidities, severity of the cancer stage, processes of care, and structure of care to the outcomes of care. We will focus on determinants of specific segments of the population not receiving quality cancer care as a means to begin the exploration of ways to remove barriers to quality care. The specific goals of the RAND-UCLA PDCR site application is to study variations and process-outcome links within the diverse setting of Los Angeles County using a population-based sample, and also to study the role of variations in structure of care for understanding variations in processes and outcomes of care for lung and colorectal cancer patients.